5,5- and 5,6-Membered Spirocyclic Indolinone Hit-Finding Libraries

Research Article Cite This: ACS Comb. Sci. 2019, 21, 528−536

pubs.acs.org/acscombsci

5,5- and 5,6-Membered Spirocyclic Indolinone Hit-Finding Libraries Peter Meier,* Nicole Battaglia, Peter Ertl, and Bernard Pirard Global Discovery Chemistry, Novartis Institutes for Biomedical Research, CH-4056 Basel, Switzerland

Downloaded via UNIV OF SOUTHERN INDIANA on July 18, 2019 at 13:50:31 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: The production of two libraries based on spirocyclic indolinones is described. These libraries were selected from numerous spirocyclic indolinone scaffolds with a library evaluation procedure used routinely at Novartis, based on computed physicochemical properties and measured properties of prototype compounds. The library production yielded 176 and 428 compounds that could be isolated in sufficient amounts and purities based on two closely related scaffolds. The novelty and diversity analysis of these libraries shows their complementarity to the chemical space covered by the structures of the PubChem database. KEYWORDS: spirocyclic compounds, combinatorial chemistry, hit finding libraries

■

INTRODUCTION Combinatorial approaches to drug discovery necessarily begin with hit-finding libraries, the quality of which are vital to the overall success of the enterprise. Although a number of novel tag-based technologies, such as DNA-encoded libraries1 or the Peptidream approach,2 have revolutionized the identification of lead structures for new and demanding targets, untagged compounds are still important for many applications, such as phenotypic or cellular assays. Here, we describe an example of an enrichment process applied to a large compound collection at Novartis,3 focusing on the evaluation of scaffold families to arrive at an optimized small library for initial hit-finding testing. We describe the preparation of spirocyclic compounds, which are especially attractive since they are rigid and, therefore, very much define the conformation of the structures.

■

Among the different unprecedented potential libraries (Figure 1), those derived from the smaller variants 3 and 4 (with R1 = R2 = H, molecular weights = 257 and 271 g/mol, respectively) were preferred, allowing a broader range of substituents to be installed while keeping the total molecular mass in the druglike range of 10 μM on 3A4BFC, 1A2CEC, E2D6AMMC, 2.8 μM on E2C19CEC, 8.5 μM on 3A4DBF),11 and no HERG channel activity (IC50 > 30 μM).12 The production of the library was estimated based on the prototype synthesis as straightforward requiring only a short development time ( 10 μM on 3A4BFC, 1A2CEC, E2D6AMMC, 1.8 μM on E2C19CEC, 0.6 μM on 3A4DBF),11 and no HERG channel activity (IC50 > 30 μM).12 The production of the library was estimated based on the prototype synthesis as straightforward requiring only a short development time ( 6.5 μmol, min purity = 85% [UV based]). The last step (the cyclization under microwave irradiation) was performed in our proprietary microwave device, which enabled the high-throughput library production.14 Here again, the classical synthesis of the prototype product 4{1,1,1} performed better than the high throughput library production conditions: 25% product could be isolated in the parallel synthesis (4{2,1,1}, the enantiomeric antipode), whereas the single compound preparation resulted in 43% product. The average yield of the isolated products in the library production was 30%. Novelty and Diversity Analysis of the Library. To analyze the novelty, diversity, and chemical space coverage of molecules in the spirocyclic library we compared it with molecules from the PubChem database.15 PubChem is the largest publicly available chemical database containing

Furthermore, the minimum energy conformation of both prototype compounds show a good overlay (Figure 4d-2) with the Cambridge Crystallographic Database (CSD) entry OSULEU (Figure 4d-1). In other words, both bicyclic prototype systems are potential replacements of the tricyclic core of the CSD entry OSULEU. This latter and the spirocyclic compounds described herein are closely related to the natural product Spirotryprostatin B.13 Accordingly, these structures are all “natural-product-like” and, as demonstrated above, conformationally highly predefined with, therefore, a superior potential for the discovery of novel promising biologically active compounds. In conclusion, the evaluation of both spirocycles did give green light for the library production. Library Production. As done generally before the library syntheses, all diversity, properties and availability based selected building blocks were tested in one reaction sequence, and only the well performing building blocks (i.e., product isolated with a yield >20% and a purity >85% after our standard reverse phase purification, see Supporting Information) were included in the library (positional scanning). This did lead to the building block selection for the 5,5-spirocycles shown in Figure 5 The library productionas described in the Supporting Informationresulted in a total of 176 products of the 5,5spirocyclic library, which did fulfill our criteria for submission to the screening deck (i.e., amount > 6.5 μmol, min purity = 85% [UV based]). The classical individual synthesis of the prototype product 3{1,1,1} performed better than the high throughput library 533

DOI: 10.1021/acscombsci.9b00057 ACS Comb. Sci. 2019, 21, 528−536

Research Article

ACS Combinatorial Science

Figure 6. Building blocks used for the 5,6-spirocyclic library.

molecules from scientific literature, patents, commercial compound providers, and other sources. The PubChem database contains currently over 97 million unique structures. The comparison has been done on the scaffold level (where scaffold was defined as the central ring part of the molecules that remains after removal of all nonring substituents). The spirocyclic library is based on 139 unique scaffolds what shows well its good diversity. Only one of these scaffolds is present within over 7 million scaffolds from the PubChem. This clearly shows that the molecules from the present spirocyclic library cover a novel, yet unexplored, area of chemical space. This may be documented also by Figure 7, where the position of the spirocyclic scaffolds is shown on the background of the PubChem scaffolds.

For this analysis, the scaffolds were characterized by the Scaffold Keys, descriptors developed specifically for the characterization of molecular scaffolds.16 For visualization, the multidimensional Scaffold Keys were reduced to two dimensions by principal component analysis,17 and the first two components were displayed. The major contributions to the first principal component are descriptors characterizing the scaffold size and to the second component are descriptors characterizing the scaffold complexity (branching and the number of heteroatoms). The occupancy of scaffold space by PubChem molecules is indicated by shading: the more densely occupied areas are darker. Also, this analysis confirms that the spirocyclic library occupies an interesting, relatively empty area of chemical space. 534

DOI: 10.1021/acscombsci.9b00057 ACS Comb. Sci. 2019, 21, 528−536

Research Article

ACS Combinatorial Science

Figure 7. Chemical space covered by the spirocyclic library (red points). The x-axis represent roughly scaffold size and the y-axis scaffold complexity (see the main text for details). The background shading represents density of chemical space in that area.

2H, ArH), 7.44−7.26 (m, 4H, ArH), 7.17−7.08 (m, 1H, ArH), 6.85 (d, J = 7.7 Hz, 1H, ArH), 6.56 (td, J = 7.6, 1.0 Hz, 1H, ArH), 6.25 (d, J = 7.5 Hz, 1H, ArH), 4.85 (d, J = 14.1 Hz, 1H, CHHPh), 4.63 (d, J = 14.1 Hz, 1H, CHHPh), 4.43 (dd, J = 11.1, 5.9 Hz, 1H, CH), 3.59 (d, J = 11.4 Hz, 1H, CH-iPr), 2.83 (dd, J = 14.1, 11.2 Hz, 1H, CHH), 2.46−2.31 (m, 1H, CHMe2), 2.19 (dd, J = 14.1, 5.9 Hz, 1H, CHH), 1.29 (d, J = 6.2 Hz, 3H, CH3), 0.50 (d, J = 6.7 Hz, 3H, CH3); minor diastereomer 8.07 (br s, 7.56, 1H, NH), 7.49−7.44 (m, 2H, ArH), 7.44−7.26 (m, 4H, ArH), 7.17−7.08 (m, 1H, ArH), 6.97 (d, J = 7.7 Hz, 1H, ArH), 4.76 (d, J = 14.5 Hz, 1H, CHHPh), 4.71 (d, J = 14.5 Hz, 1H, CHHPh), 4.58 (t, J = 8.3 Hz, 1H, CH), 4.01 (d, J = 8.2 Hz, 1H, CH-iPr), 1.98−1.78 (m, 3H, CHMe2) 1.13 (d, J = 6.7 Hz, 1H, CH3), 0.60 (d, J = 6.5 Hz, 3H, CH3); ratio of disastereomers = 1:3. 13C NMR (100 MHz, MeOD): δC major diastereomer 179.8, 173.3, 157.4, 141.0, 136.3, 130.6, 129.0, 128.4, 128.1, 127.8, 123.1, 122.3, 110.1, 75.0, 60.5, 58.2, 42.3, 40.4, 26.6, 22.2, 17.4; minor diastereomer 180.6, 173.6, 161.4, 141.5, 136.2, 129.1, 128.8, 128.3, 127.7, 127.4, 124.7, 122.1, 110.2, 70.5, 63.0, 59.5, 42.2, 39.2, 30.7, 19.8, 18.3. HRMS m/z Calcd for C23H23N3O3 [M + H]+: 390.18122. Found: 390.18124. Procedure for the Preparation of (3S,6′S,8a′S)-2′Benzyl-6′-isopropyl-2′,3′,8′,8a′-tetrahydro-1′H-spiro[indoline-3,7′-pyrrolo[1,2-a]pyrazine]-1′,2,4′(6′H)-trione 4{1,1,1}. Methyl(2′S,3S,5′S)-2′-isopropyl-2-oxospiro[indoline-3,3′-pyrrolidine]-5′-carboxylate 2{1,1} (66 mg, 0.23 mmol) was dissolved in DCM (4 mL). Triethylamine (38 μL, 0.27 mmol, 1.2 equiv) and chloroacetyl chloride (20 μL, 0.253 mmol, 1.1 equiv) were added subsequently. The resulting reddish reaction mixture was stirred for 2 h at rt. The volatiles were evaporated, and the residue was dissolved in EtOAc (100 mL) and washed with water (2 × 30 mL), 1 M KHSO4 (20 mL), and brine (30 mL). The organic layer was dried (sodium sulfate), filtered, and evaporated to give a brownish solid (173 mg). This solid was dissolved in MeOH (4 mL), and benzylamine 15{1} (51 μL, 0.47 mmol, 2 equiv) was added. The reaction mixture was refluxed for 20 h. After it was cooled to rt, the volatiles were evaporated. The residue was purified by column chromatography (silicagel, DCM/MeOH 19:1) to give

In conclusion, we exemplified the evaluation process for compound libraries applied at Novartis with different spirocyclic indolinone based scaffolds. The calculated and measured properties led to the selection and production of the described 5,5- and 5,6-membered libraries. The novelty and diversity analysis of the 176 and 428 compounds isolated for these libraries did show the complementarity of these derivatives compared to the chemical space covered by the PubChem structures. The compounds were added to the Novartis compound collection, which is continuously used in hit-finding campaigns.

■

EXPERIMENTAL PROCEDURES The methyl-2′-alkyl-2-oxospiro[indoline-3,3′-pyrrolidine]-5′carboxylates were prepared according to literature.5 Procedure for the Preparation of (3S,5′S,7a′S)-2′Benzyl-5′-isopropyl-7′,7a′-dihydro-1′H,5′H-spiro[indoline-3,6′-pyrrolo[1,2-c]imidazole]-1′,2,3′(2′H)-trione 3{1,1,1}. Methyl (2′S,3S,5′S)-2′-isopropyl-2-oxospiro[indoline-3,3′-pyrrolidine]-5′-carboxylate 2{1,1} (100 mg, 0.30 mmol) was dissolved in dry DCM (10 mL). Benzylisocyanate 12{1} (40.4 μL, 0.33 mmol, 1.1 equiv) was added, and the reaction mixture was stirred for 17 h at rt. The volatiles were evaporated under vacuum; the residue was dissolved in EtOAc (150 mL) and washed with 1 N HCl (2 × 30 mL) and brine (30 mL). The organic layer was dried (sodium sulfate), filtered, and evaporated. The obtained brownish foam was dissolved in AcOH (3 mL). 3 N HCl (3 mL) was added, and the resulting brownish-red reaction mixture was stirred for 2 d at rt. AcOH was evaporated under vacuum (water bath 40 °C, at 40 mbar), the remaining solution was poured into sat. aq. NaHCO3 (60 mL) and extracted with TBME (3 × 50 mL). The combined organic layers were washed with sat. aq. NaHCO3 (3 × 30 mL), dried (sodium sulfate), filtered, and evaporated. The crude was purified by preparative TLC (2% MeOH in DCM; silicagel plate 1 mm × 20 cm × 20 cm; Rf of the product = 0.25) to give the desired product 3{1,1,1} as a white powder (74 mg, 0.18 mmol, 61%). 1 H NMR (400 MHz, MeOD): δH major diastereomer: 7.95 (br s, 1H, NH), 7.95 (br s, 1H, NH), 7.56 (dd, J = 7.5, 1.9 Hz, 535

DOI: 10.1021/acscombsci.9b00057 ACS Comb. Sci. 2019, 21, 528−536

Research Article

ACS Combinatorial Science

(6) Lipinski, C. A. Lead- and drug-like compounds: the rule-of-five revolution. Drug Discovery Today: Technol. 2004, 1 (4), 337−341. (7) Ertl, P.; Muehlbacher, J.; Rohde, B.; Selzer, P. Web-based cheminformatics and molecular property prediction tools supporting drug design and development at Novartis. SAR QSAR Environ. Res. 2003, 14 (5−6), 321−328. (8) Alsenz, J.; Kansy, M. High throughput solubility measurement in drug discovery and development. Adv. Drug Delivery Rev. 2007, 59 (7), 546−567. (9) Wohnsland, F.; Faller, B. High-throughput permeability pH profile and high-throughput alkane/water log P with artificial membranes. J. Med. Chem. 2001, 44 (6), 923−930. (10) Avdeef, A.; Bendels, S.; Di, L.; Faller, B.; Kansy, M.; Sugano, K.; Yamauchi, Y. Parallel artificial membrane permeability assay (PAMPA)-critical factors for better predictions of absorption. J. Pharm. Sci. 2007, 96 (11), 2893−2909. (11) Li, G.; Huang, K.; Nikolic, D.; van Breemen, R. B. Highthroughput cytochrome P450 cocktail inhibition assay for assessing drug-drug and drug-botanical interactions. Drug Metab. Dispos. 2015, 43 (11), 1670−1678. (12) Gillie, D. J.; Novick, S. J.; Donovan, B. T.; Payne, L. A.; Townsend, C. Development of a high-throughput electrophysiological assay for the human ether-a-go-go related potassium channel hERG. J. Pharmacol. Toxicol. Methods 2013, 67 (1), 33−44. (13) Cui, C.-B.; Kakeya, H.; Okada, G.; Onose, R.; Ubukata, M.; Takahashi, I.; Isono, K.; Osada, H. Tryprostatins A and B, novel mammalian cell cycle inhibitors produced by Aspergillusfumigatus. J. Antibiot. 1995, 48, 1382−1384. (14) Chamoin, S.; Moser, M.; Roth, H.-J.; Zahnd, B. Automated device comprising microwave irradiation for synthesis of organic compound libraries. PCT Int. Appl. WO2008000804, 2008. (15) Kim, S.; Chen, J.; Cheng, T.; Gindulyte, A.; He, J.; He, S.; Li, Q.; Shoemaker, B. A.; Thiessen, P. A.; Yu, B.; Zaslavsky, L.; Zhang, J.; Bolton, E. E. PubChem 2019 update: improved access to chemical data. Nucleic Acids Res. 2019, 47 (47), D1102−D1109. (16) Ertl, P. Intuitive ordering of scaffolds and scaffold similarity searching using scaffold keys. J. Chem. Inf. Model. 2014, 54, 1617− 1622. (17) Hastie, T.; Tibshirani, R.; Friedman, R. The Elements of Statistical Learning; Springer: New York, 2017.

the product 4{1,1,1} as a white solid (40 mg, 0.10 mmol, 43%). 1 H NMR (400 MHz, MeOD): δH 7.39−7.27 (m, 7h, ArH), 7.07 (td, J = 7.6, 1.0 Hz, 1H, ArH), 6.94 (d, J = 7.8 Hz, 1H, ArH), 4.98−4.90 (m, 2H, CH and CHH), 4.41 (d, J = 14.8 Hz, 1H, CHH), 4.30 (d, J = 17.1 Hz, 1H, CHH), 3.94 (d, J = 5.4 Hz, 1H, CHH), 3.86 (d, J = 17.1 Hz, 1H, CHH), 2.72 (dd, J = 13.4, 10.6 Hz, 1H, CHH), 2.34 (dd, J = 13.4, 7.3 Hz, 1H, CHH), 2.26−2.17 (m, 1H, CH), 0.88 (d, J = 6.9 Hz, 3H, CH3), 0.73 (d, J = 6.8 Hz, 3H, CH3). 13C NMR (100 MHz, MeOD): δC 181.4, 169.2, 168.2, 142.0, 136.2, 128.9, 128.53, 127.8, 127.7, 127.1, 125.5, 122.0, 109.8, 66.8, 57.1, 56.0, 51.5, 33.5, 30.9, 21.4, 18.0. HRMS m/z Calcd for C24H25N3O3 [M + H]+: 404.19687. Found: 404.19681.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acscombsci.9b00057.

■

Experimental procedures for the library preparation and specifications for all final compounds (PDF)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Peter Meier: 0000-0002-5160-183X Peter Ertl: 0000-0001-6496-4448 Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS We thank Felix Thommen for the purification of all library compounds by preparative HPLC. REFERENCES

(1) Mannocci, L.; Leimbacher, M.; Wichert, M.; Scheuermann, J.; Neri, D. 20 years of DNA-encoded chemical libraries. Chem. Commun. 2011, 47, 12747−12753. (2) (a) Reid, P. C.; Goto, Y.; Katoh, T.; Suga, H. Charging of tRNAs using ribozymes and selection of cyclic peptides containing thioethers. Methods Mol. Biol. 2012, 805, 335−348. (b) Murakami, H.; Kawakami, T.; Reid, P. M.; Sasaki, T. (University of Tokyo; Peptidream, Inc.) US Patent0272964, 2016. (3) Jacoby, E.; Schuffenhauer, A.; Popov, M.; Azzaoui, K.; Havill, B.; Schopfer, U.; Engeloch, C.; Stanek, J.; Acklin, P.; Rigollier, P.; Stoll, F.; Koch, G.; Meier, P.; Orain, D.; Giger, R.; Hinrichs, J.; Malagu, K.; Zimmermann, Z.; Roth, H.-J. Key Aspects of the Novartis Compound Collection Enhancement Project for the Compilation of a Comprehensive Chemogenomics Drug Discovery Screening Collection. Curr. Top. Med. Chem. 2005, 5, 397−411. (4) Miyake, F. M.; Yakushijin, K.; Horne, D. A. Preparation and synthetic applications of 2-halotryptamines: synthesis ofelacomine and isoelacomine. Org. Lett. 2004, 6 (5), 711−713. (5) Miyake, F. M.; Yakushijin, K.; Horne, D. A. Preparation and synthetic applications of 2-halotryptophan methyl esters: synthesis of spirotryprostatin B. Angew. Chem., Int. Ed. 2004, 43, 5357−5360. 536

DOI: 10.1021/acscombsci.9b00057 ACS Comb. Sci. 2019, 21, 528−536